This cluster-randomized school-based study will examine the effectiveness of a teen dating violence (TDV) and sexual violence (SV) prevention program for preventing serious violence perpetration among middle school students who have been exposed to violence. ?Expect Respect? (ER) gender-specific support groups are for youth with history of exposure to violence including witnessing domestic violence, loss of loved one to homicide, TDV, and SV victimization, who are thus at elevated risk for TDV/SV and other violence perpetration. ER is a theory- and research-informed program intended to alter norms condoning violence and rigid gender expectations that foster violence perpetration, promote bystander intervention (i.e., give youth skills to interrupt disrespectful and abusive behaviors they witness among peers), and reduce TDV/SV perpetration. Implemented by trained facilitators, the 24-session curriculum focuses on gender equity and respect, recognizing abuse and impact on self and others, regulating emotions, skills for healthy relationships, and becoming active proponents for safe and healthy relationships. The study focuses on middle school students because the prevalence of witnessing and experiencing harmful behaviors is already high in middle school as are gender inequitable attitudes, suggesting that efforts to support vulnerable students and to shift attitudes and social norms condoning violence are urgently needed during this developmental period. In a recent controlled evaluation conducted by the Centers for Disease Control and Prevention (CDC), both boys and girls demonstrated significant reductions in reactive and proactive aggression compared to control participants. Given reduction in aggression for both genders, we propose to study this violence prevention program, created to target TDV/SV, for its effectiveness in preventing violence perpetration more broadly including weapon carrying and physical fighting. While the goal is to prevent perpetration of serious violence, we also anticipate reductions in precursors of such violence including sexual harassment, homophobic teasing, and reactive and proactive aggression towards peers. Control schools will provide student support services as usual. Via a 2- arm cluster-randomized trial in 36 middle schools (baseline N =1080), we will evaluate the effectiveness of this support group-based prevention program for vulnerable students on serious violence perpetration (primary outcome), weapon carrying (secondary), positive bystander behaviors (secondary), and suicidality (exploratory) (Aim 1). One year after baseline, we will assess these same outcomes compared to control students (Aim 2). We will explore whether demographics, pre-intervention risk and protective factors (e.g., history of violence, school connectedness), and implementer characteristics may interact with the intervention to predict changes in outcomes of interest (Aim 3). This study is significant because we have too few evidence- based violence prevention programs for any age, none tailored for youth already exposed to violence, and none that directly address multiple violence outcomes including serious and lethal violence.